Grid-scale energy storage is the lesser-publicized half of the clean energy story. As solar and wind farms scale up, so does the grid’s need to put electricity on layaway for those nights and cloudy and windless days when solar and wind farms lie fallow. Storing electric power via flywheels, compressed air, superconductors, pumped water reservoirs, thermal storage, hydrogen gas, and even rocks on railcars are methods being researched—and in some cases, commercially prototyped today.
But ARPA-E, the U.S. Department of Energy’s technology incubator, retains a strong focus on the familiar electrochemical battery as a likely backbone of an increasingly solar and wind-powered electric grid of the future.
A report published by ARPA-E earlier this year outlines the $85 million in R&D funding it’s invested in battery-based grid storage since 2009. The report details the agency’s grid-scale battery plans, including projected system costs, which they say should eventually drop by at least an order of magnitude compared with 2010. That, say ARPA-E researchers, would allow them to become viable commercial players in the years ahead.
According to Eric Rohlfing, ARPA-E’s Deputy Director for Technology, some technologies they’ve invested in are headed that way. But he notes that the agency backs portfolios of projects, and is not in the business of picking single winners in any category. “One of the things that we like to say we do at ARPA-E is we provide technological options,” Rohlfing says. “As much as we love many of these projects, for me to look into a crystal ball and say, ‘This particular chemistry, this particular flow battery will be the answer,’ I think is premature.”
That said, though, Rohlfing called the aforementioned report a sampler of some of the 73 ARPA-E–supported grid storage projects that appear poised for commercial opportunities in the months and years ahead. (IEEE Spectrum discussed some of these projects in a previous story.)
One of these grid-scale battery companies, Portland, Ore.-based Energy Storage Systems (ESS), now has its batteries installed in an Army Corps of Engineers deployment and at a winery in Napa, Calif. Its technology is an example of a new approach to a promising grid storage idea: the flow battery.
A flow battery is like a melding of a fuel cell and a conventional battery. A liquid electrolyte flows in both the cathode and anode sides of the cell; they’re separated by a membrane. The battery’s capacity can be easily expanded by adding more reservoirs of electrolyte. If the chemistry and engineering is done right, the battery should have enough capacity to handle grid-scale power needs while still remaining neither expensive, nor toxic, nor volatile.
Of course, that’s easier said than done.
“We’re seeing a lot of potential in these flow battery systems,” Rohlfing says. “You have two large tanks that you can store a lot of energy in. And you can ramp up that energy scaling quite easily. And your active medium is a small part of that, that the reactive parts are flowing through. When we first started, the state of the art was based on all vanadium flow batteries—which were and still are very expensive, because of the vanadium. All of our projects have looked at ways of lowering that cost. Energy Storage Systems originally started with vanadium and has shifted to an all–iron-chloride approach. And iron is dirt cheap… or rather, rust cheap.”
Bill Sproull, VP of Business Development and Sales at ESS, says the price of the company’s iron-based electrolyte is an order of magnitude cheaper than the vanadium-based electrolytes he’s studied. “The cost of our electrolyte today is somewhere on the order of $15 per kilowatt-hour. And my understanding of the cost of the vanadium electrolyte is somewhere in the $150 per kilowatt-hour range,” he says. “Given that, you’ve got to still build the rest of the battery out of low-enough-cost materials that you’re not going to reverse that situation.”
Julia Song, the company’s co-founder and CTO, says that at least one other company and two universities are also working on their own iron-based flow battery chemistry. And no doubt all competitors are facing up to one of the key challenges for an iron-based electrolyte, she says. Namely, the battery’s positive and negative electrolytes perfer t opareat at different pH levles for optimal performance. So there need to be some clever chemistry and chemical engineering efforts to keep electroyltes’ pH separate and stable during operation.
Sproull and Song both say that this is a tricky, but not impossible problem of optimization. And it ultimately needs to be solved only once. This is why they say ARPA-E’s support has been so important for the company and for the technology.
In 2012, ARPA-E awarded ESS $2.1 a million grant for what it called “transformational energy storage projects.”
“That really changed everything for us,” Song says. “We were able to get the people, get the space, do the research, and build early prototypes that we’re good at for two-and-a-half years—and really focus on technology development without worrying about raising money. That made a huge difference.”
Since then, ESS, armed with its more mature technology, has been able to raise venture capital, says Sproull. And now the company projects that it’ll be volume manufacturing its flow battery system within a year.
Scottsdale, Ariz.-based Fluidic Energy has, with ARPA-E support, developed a zinc-air battery that serves a different need than ESS’s iron flow batteries. Says the company’s CTO Ramkumar Krishnan, its zinc-air batteries have been developed to supplant lead-acid batteries and diesel generators—especially in countries with developing electricity grids (e.g. rural electrification) or for long-duration critical backup applications (e.g. cellphone towers), where power reliability needs can be met whenever sporadic power sources drop out.
Zinc-air batteries have been around for decades, powering hearing aids primarily. But it was with ARPA-E support that Fluidic tweaked the materials and the design to make its zinc-airs rechargeable.
The ARPA-E report described the multiple challenges Fluidic faced along the path to developing the technology. Fluidic, ARPA-E said, “focused on developing a battery design using an electrolyte based on ionic liquids. Ionic liquids are salts that are liquid at the battery operating temperature, delivering ionic conductance while maintaining substantial electrical insulation. The team developed chemistries that have negligible evaporation, are stable in the presence of oxygen, and do not absorb water over the cell operating voltages, and include additives that interact favorably with the zinc and air electrodes.”
ARPA-E is telling its grantees that, in order to be cost competitive in the commercial marketplace, they need to target battery benchmarks of $100 per kilowatt hour, with at least 5000 cycles (i.e. 10 years daily operation) and 80-percent or greater efficiency in cycling from charging to discharge.
Fluidic’s Krishnan says that Fluidic is competitive in all three of those categories. For instance, he says, “Today we’re four to five times over the life of lead-acid batteries. We’re providing a five-year warranty in telecom applications, where typically their [lead acid] batteries are replaced every 18 months.”
Krishnan says that, as with ESS, ARPA-E provided crucial support at an important time in the Fluidic technology’s development. “ARPA-E plays that really fine balance between pure R&D funding that is provided by national endowments and national labs, versus ventures that are funding something that has quick payback and low risk for technology to market entrance,” he says.
“Just like how in [President] Kennedy’s time, going to the moon was a dream. It was made a reality by setting out some bold visions and providing a means to achieve that. Similar to that, ARPA-E is breaking some new ground by allowing some bold technologies to be able to push the forefront, [letting companies] make that viable in a short timeframe, and take it to the next level where it becomes attractive for investors or the public to fund that further.”
This post was corrected on 7 November 2016 to better characterize the technologies of ESS and Fluidic Energy.